VOXL 3 - Maximizing Image Processing per Gram

Thermal longwave infrared (LWIR) capabilities are essential for enhancing the operational effectiveness of small Unmanned Aircraft Systems (UAS) in the US Army. ModalAI proposes to develop an advanced thermal imaging module by upgrading its existing VOXL 2 system to the new VOXL 3 platform. The VOXL 3, featuring the Qualcomm QCS8550 chipset, will triple the deep neural network processing power to 45 TOPS and support three concurrent 18-bit Image Signal Processors (ISPs). This upgrade will enable unprecedented target recognition and superior image quality while maintaining a lightweight form factor of 5–10 grams, including the lens. Our solution leverages proven technologies and a Modular Open Systems Approach (MOSA) to ensure compatibility and scalability. The enhanced thermal processing capabilities will be deployed at the soldier level, significantly improving situational awareness and mission effectiveness. ModalAI's proposal aligns with the Army's goals of reducing weight, increasing performance, and maintaining low unit costs for thermal imaging systems in UAS. With a strong financial standing, extensive experience in Qualcomm processor-based designs, and a successful track record in both commercial and defense sectors, ModalAI is well-positioned to deliver a cutting-edge solution. This proposal outlines our comprehensive plan to develop, test, and demonstrate the upgraded thermal imaging module, ensuring robust performance and reliability for Army applications.